Comparative visual analytics for multi-modal single-cell and spatial omics data

Abstract

Advances in single-cell and spatial omics technologies have transformed biological research by enabling increasingly detailed characterization of gene expression, cell-type organization, and tissue structure. However, the rapid growth and complexity of these datasets have created new challenges for their exploration and interpretation. Conventional computational workflows often rely on predefined analyses and static visual outputs, making it difficult for researchers to iteratively investigate complex relationships and refine their questions.

To support the interactive exploration and comparison of large-scale biological data, this work develops visual analytics frameworks that integrate computational analysis with interactive visualization. Using design study methodologies in close collaboration with domain experts, the frameworks support the exploration of complex relationships within and across heterogeneous datasets while allowing researchers to dynamically filter, compare, and investigate patterns in their data. The resulting research has been translated into a tangible software solution that brings these analytical and visualization capabilities together in a modular environment, connecting complementary views and analytical methods while preserving biological context.

The frameworks and software system demonstrate how interactive visual analytics can make the analysis of large and complex biological datasets more flexible, interpretable, and iterative, supporting researchers in generating, refining, and validating hypotheses during data exploration.